The advent of moored technologies for harmful algal bloom (HAB) detection with real-time telemetry facilitates opportunistic and adaptive sampling of key bloom processes that have as yet evaded us in traditional seagoing survey mode. We propose to use data streams from Environmental Sample Processor (ESP) moorings to mobilize rapid-response surveys during key phases of two types of HABs in the Gulf of Maine: Alexandrium fundyense and Pseudo-nitzschia spp. For A. fundyense, sampling will focus on bloom decline, as the mechanisms that terminate these blooms are as yet undocumented in the field. The dynamics of Pseudo-nitzschia spp. remain largely uncharacterized in this region, and rapid response sampling will accordingly be targeted amongst the various phases of initiation, development, and decline. Laboratory experiments will be conducted with Pseudo-nitzschia isolates collected from the Gulf of Maine to measure their vital rates and dependencies on environmental conditions. Knowledge gained from both field and laboratory studies will be incorporated into models of bloom dynamics to improve predictive capabilities.